The cerebral degeneration of Alzheimer's disease (AD) results in the majority of the cases of dementia in the aging human population. One of the early events in the pathogenesis of AD appears to be the deposition of beta-amyloid protein (B/A4), a 39-42 amino acid proteolytic cleavage product of the much larger amyloid precursor protein (betaAPP). Beta amyloid protein in AD is associated with the serin protease inhibitor alpha1-antichymotrypsin (ACT). Beta amyloid protein deposition has been positively correlated with the astrocytosis of AD and astrocytes are also a source of B/A4 and ACT. Our preliminary studies indicate the presence of critical regulatory elements within a discrete region of 5' flanking DNA of the human GFAP (hGFAP2) gene that can be used to target the expression of heterologous genes to astrocytes in vivo. We plan to test the in vivo capability of astrocytes to be the primary instigators of amyloidogenesis by manipulating astrocyte gene expression in transgenic mice. The hGFAP2 promoter sequence will be used to direct the over expression of betaAPP751, betaAPP751/Goate mutant and betaAPP-C104 forms of the betaAPP gene as well as ACT to the astrocytes of transgenic mice. Transgenic mice expressing the various betaAPP constructs will additionally be crossed with ACT expressing mice to create double transgenics. All mice will be monitored for transgene transmittance via PCR and Southern hybridization, and transgene expression via Northern hybridization. Sacrificed animals will be analyzed with appropriate routine and special histopathology (immunocytochemistry), Thioflavin S) and electron microscopy.